Recent data have shown that the human genome can be objectively parsed into haplotype blocks: sizeable regions over which only a few common haplotypes capture the vast majority of sequence variation in the population. Unraveling the inherited contribution to human disease will require systematic assessment of these haplotypes (and the SNPs that track with them in the population) for association to disease phenotypes. To enable such studies we propose the creation of a genome-wide map of human haplotypes, requiring the genotyping of at least 850,000 high frequency variants in at least 188 individuals from Europe, Asia and Africa. Our Center proposes to define haplotype blocks over 24% of the human genome, developing up to 370,000 SNP genotyping assays and performing up to 71 million genotypes. SNPs will be selected and genotypes obtained in a regional, hierarchical design, with up to three rounds of genotyping guided by the haplotype patterns observed in previous rounds. Haplotype blocks will be defined using an empirically validated suite of methods described in a recent publication from our group (Gabriel et al., "The Structure of Haplotype Blocks in the Human Genome, Science, May, 2002). The proposal is based on our demonstrated experience with MALDI-TOF genotyping, with which we have produced over 7,000,000 genotypes in the last 12 months, achieving pass rates of SNPs of 80% and accuracy of >99.6%. While confident that this technology will support and is suitable for the HapMap project, we will continue to evaluate new genotyping methods to ensure the highest quality, efficiency and lowest cost for HapMap. As the data produced by HapMap will shed light on human history and genetic diversity, we also propose to create an Education and Outreach Program to share these developments with our colleagues and engage our communities. The haplotype map project will catalyze human genetic studies of unprecedented power and scope, promising increased insight into disease risk in the population, and ultimately, improved clinical care.